The proposed research has three closely related objectives. The first is to develop and study different models that describe the regulatory interaction between genes. The second objective is to develop fast algorithms based on these models for detecting regulatory interactions between the genes inferred from the expression array data, i.e. for reverse engineering of the network architecture from the activity profiles of genes in the system. This includes the development of parallel computational methods for analyzing gene sequences at a much higher speed than the sequential methods. Inference about the causal relationships among genes should generate hypotheses for further experiments. The third objective is to use these methods to study the global signal circuitry of different organisms and cell types and to explain how specific generic lesions serve to reprogram the wiring diagram of cellular signaling pathways in each of the constituent cell types so as to manifest cancer and other diseases.